Sanity check of TVR cooling system

[dnb]

I am about to put my rebuilt engine back in the TVR, so it's time to think about the cooling system.

I have sketched the existing system on the left, and what I think it ought to be like on the right. I have neglected to include things like overflows and the throttle body heater for clarity. The colours indicate flow direction.



I was never really keen on the TVR implementation of the cooling system since it didn't have a proper thermostat bypass - it relied on the heater which had a stop valve in it (until I redesigned the heater to not have the valve...) and it's always seemed like the header tank would be better on the other side of the radiator.

So I took some inspiration from the Thor P38 cooling design. If I drop the existing thermostat and use the remote stat from a P38 and add some bypass pipes then I end up with something like the drawing on the right.

Is it worth making the changes to make a more robust system? And is the original design as bad as I think it might be? The changes aren't much effort to make.

Thanks for any insight.

[ChrisJC]

Is the one on the right not exactly the same as the P38 cooling system?

My only reservation would be that with the bottom-hose 'stat, there are no choices on stat temperature that I am aware of.

I have a P38, on which the cooling system works fine. I also did a very similar job on my Series Landie, and that also works just fine. (albeit with an expansion tank rather than a header tank)

Chris.

[dnb]

Yes - schematically it should be identical to a P38 system (apart from any errors I introduced). In terms of implementation on the TVR it will end up with minor differences (e.g. things on the other branch of a T)

I wasn't aware there weren't a range of stats so that's good info - thanks. I was planning to use a standard stat for this anyway so it's not a problem unless I discover the TVR wants to run cooler - according to the RAVE documentation, the stat should be OK for my application.

[SuperV8]

Mine looks the same as the one on the right. I used a thermostat housing from an MGF or freelander as they have a lower temp opening than the range rover housing. They are also pressure relieving so open with pressure/flow or temperature which helps save head gasket failures apparently. They have less inputs than the range rover stat but I just used a Y joint to connect up heater return.

There are different colours for different temperatures. Mines grey. Works fine, sits between 85 and 90 when driving.
something like this one I think:
http://www.rimmerbros.co.uk/Item--i-PEL ... wwodfK0AMQ

Tom.

[Darkspeed]

Only comment I would make is that the small air bleed to the header should be from the highest point of the system and the highest pressure.

Which it probably is with the stat in the return and a high rad as in the P38.

In a TVR it will not be the rad - With a standard manifold stat position it will not be the rad.

Highest point highest pressure for the feed - as the bottom of the header is at lowest point and under suction. This give a high operating diffential in the header to ensure full dearation of the fluid.

Andrew

[dnb]

Thanks. More good info.

I can't see a reason to not try the remote stat scheme since I need to change the stat regardless. I never liked the concept of using the heater as a bypass given the strangled route of the hoses.

I will draw up a revised schematic and add some pictures.

[spend]

I prefer the Cerb/Citroen/Alfa stat to the P38. It's metal and can just bypass the Rad.

Putting an expansion tank up on the plenum worked really well, with vents from the swirl tank and extra bleeds in the manifold (old throttle take off + 2 tapped into head water way at rear of manifold). For extra bleeding on the Griffs damn rad this was also linked into the venting using small bore hose and banjo into the rad bleed screw (now redundant). Heater wise I just took the feed from the manifold as usual and returned it into a hosetail drilled & tapped into the nose of the waterpump.

Most of that was to suit using water/oil cooler, but I see no reason it wouldn't work well without one. Bleeding was now a doddle whilst we flushed the system several times to get rid of old anti-freeze traces, just fill & it all works.

Adding a drain tap under the rad was also a boon during flushing, and will enable us to keep the expensive synth coolant in use if we ever need to drain for work.

Cliff has Pascos number, sure he'd send plenty of pictures if you want.

[dnb]

Hi Spend. Been a while... How's things going?


I decided against the Citroen stat simply because I got myself into a state of needing the extra port on the P38 stat - it made another job (nothing to do with cooling) considerably easier. Other than that, I've done much the same as you've done.

I managed to route the heater pipes under the inlet manifold to get them out of the way. It had the benefit of a cooling pipe on the right side of the rad going near the header tank lower connection amongst other things.

Anyway, pictures are usually better than words at explaining things. Here's the engine bay laid out with odds and ends of tube and spare bits of hose to see if my crazy plan will work... It should match the schematic posted earlier.

[spend]

I think the issue you may have is with the header tank height (subaru?) ~ it's not really any higher then the inlet manifold so allowing for expansion level will not ensure fluid in manifold / heads?

I made up small tank to clip on plenum, with Cliffs front mounted coil pack arrangement so room above FPR for small tank + mount for FPR.

[dnb]

The top is quite a bit higher than the manifold. The photo isn't particularly clear showing this because the rocker cover is on shims to check clearance. Granted there could alwaysbe more height! It doesn't have a standard plenum so I can't do exactlywhat you did.

[dnb]

Happiness is a TVR that doesn't overheat when sat in rush hour traffic.



The Range Rover configuration would seem to be a substantial improvement over standard operation.